Published ahead of print on August 15, 2002, doi:10.1164/rccm.200202-155OC
© 2003 American Thoracic Society
Endothelial Nitric Oxide Synthase Variants in Cystic Fibrosis Lung DiseaseChildren's Hospital, University of Essen, Essen, Germany; Department of Medicine and Channing Laboratory, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts; and Department of Epidemiology and Biostatistics, Case Western Reserve University, Cleveland, Ohio Correspondence and requests for reprints should be addressed to Dr. Hartmut Grasemann, Children's Hospital, University of Essen, Hufeland Str. 55, D-45122 Essen, Germany. E-mail: hartmutg{at}hotmail.com
Variants in the genes encoding for the nitric oxide synthases may act as disease modifier loci in cystic fibrosis, affecting both an individual's nitric oxide level and pulmonary function. In this study, the 894G/T variant in exon 7 of the endothelial nitric oxide synthase gene was related to exhaled nitric oxide and pulmonary function in 70 cystic fibrosis patients who were aged 14.8 ± 6.9 years (mean ± SD), with a FEV1 of 69.4 ± 24.8% predicted. Although there was no association between endothelial nitric oxide synthase genotypes and exhaled nitric oxide in males, nitric oxide levels were significantly higher in female cystic fibrosis patients with an 894T mutant allele, compared with female patients homozygous for the 894G wild-type allele (7.0 ± 4.4 versus 3.6 ± 1.9 parts per billion, p = 0.02). Furthermore, in female patients, colonization of airways with Pseudomonas aeruginosa was significantly (p < 0.05) less frequent when carrying an 894T mutant allele as compared with wild type. These data suggest that the 894T variant in the endothelial nitric oxide synthase gene is associated with increased airway nitric oxide formation in female cystic fibrosis patients, possibly affecting colonization of airways with P. aeruginosa.
Key Words: endothelial nitric oxide synthase • exhaled nitric oxide • cystic fibrosis • Pseudomonas aeruginosa Nitric oxide (NO) has been shown to be important in a variety of regulatory processes in the lung, including host defense, inflammation, and bronchomotor control (1, 2). There are three known isoforms of NO synthases (NOSs): inducible NOS, neuronal NOS (NOS1), and endothelial NOS (NOS3). Inducible NOS is regulated by cytokines primarily at the level of gene transcription, whereas NOS1 and NOS3 are constitutively expressed in various cell types, and their activity has been thought to be modulated primarily by Ca2+ and calmodulin-dependent mechanisms. However, it has now been established that NOS1 and NOS3 are also regulated by gene induction at multiple levels, including transcriptional control (3). NO concentrations measured in the exhaled air are highly variable but are generally increased in inflammatory lung diseases such as asthma or bronchiectasis. Patients with cystic fibrosis (CF) also have variable airway NO concentrations; however, even though CF is an inflammatory disease of the airways, NO levels are paradoxically decreased (4, 5). The underlying mechanisms for the decreased, but variable, NO concentrations in CF are incompletely understood but may reflect both differences in formation of NO and retention of NO in CF airway fluids. Of potential clinical relevance, fraction of exhaled NO (FENO) in CF has been shown to correlate positively with lung function, suggesting that the formation of airway NO is lower in patients with poor pulmonary function (4, 6). Furthermore, low FENO is associated with an increased rate of bacterial colonization of the CF airways (7). NOS3 is constitutively expressed in pulmonary blood vessels, airway epithelial cells, and neutrophils (8–10). There is increasing evidence that NOS3 contributes to NO-related physiology and pathophysiology in the airways (10–12). The NOS3 gene, which is located in chromosomal region 7q35–36, contains a number of gene variants, including a G/T polymorphism at position 894 in exon 7 (13). This 894G/T substitution, which results in an aspartate rather than glutamate at position 298 in the NOS3 protein, is of potential functional relevance as it confers stability of NOS3 to proteolytic cleavage in vitro (14). Furthermore, the activity of NOS3 has been shown to be different between sexes. Activation of NOS3 by circulating estrogen results in higher NOS3 activity in vascular endothelium of females compared with males (15). CF is an inherited Mendelian disorder of chloride channel function and is clinically characterized by chronic deterioration of pulmonary function. Because the clinical presentation of CF is not closely related to the disease-causing mutations in the CF transmembrane regulator (CFTR) gene, other factors, including naturally occurring variants in disease-modifying genes, may affect the course of CF disease (16). In this study, we analyzed the relationship of the NOS3 894G/T gene variant with airway NO concentration and colonization of airways in CF patients.
Subjects Seventy unrelated patients with CF, 4 to 40 years old, were studied. The diagnosis of CF was confirmed by repeated sweat tests with chloride concentrations of more than 60 mmol/L and by mutation analysis of the CFTR gene. All patients were recruited at the CF outpatient department of the Children's Hospital at the University of Essen. At the time of study, patients had to be without signs of acute upper airway infection or pulmonary exacerbation and without a need for intravenous antibiotics, as well as systemic or inhaled steroids for at least 2 weeks before their study visit. Pulmonary function data at the time of FENO measurements were available for 69 CF patients. The basic characteristics of the study population are displayed in Table 1.
NO Measurement FENO was measured with a chemiluminescence analyzer (NOA 280; Sievers, Boulder, CO) in accordance with published American Thoracic Society standards (17). Single-breath online measurements were performed at a constant expiratory flow of 100 ml x second-1 for the assessment of lower airway NO, as recently described (17). The mean value of three end-expiratory NO concentrations was calculated from each subject.
Genotyping
Statistics Sex was also analyzed as a binary variable. All other variables were analyzed as continuous. Quantitative variables were expressed as mean ± SEM. All quantitative variables were approximately normally distributed. Bivariate analysis used correlation coefficients, t tests, or analysis of variance to compare mean FENO across levels of 894G/T genotype. Generalized linear models (linear regression) (19) were constructed to model the effects of multiple covariates on FENO level. Consistency of genotype frequencies with Hardy-Weinberg equilibrium was tested using a chi-squared goodness-of-fit test on a contingency table of observed versus predicted genotype frequencies. Statistical significance was taken at the conventional 5% level. PC-Statistik version 2.11 (TopSoft, Hannover, Germany) and Minitab v13 (Minitab Inc., State College, PA) were used to manage and analyze data.
Total Population There were more males than females in the study sample (Table 1). Males and females did not differ by mean age (t47 = 0.87, p = 0.39), FEV1 percentage predicted (t52 = 0.33, p = 0.74), or FENO levels (t60 = 0.11, p = 0.91). A statistically significant negative correlation between age and FENO was found in male (r = -0.42, p = 0.008) but not female (r = -0.12, p = 0.53) CF patients.
Allele frequencies of the marker in the NOS3 gene were 0.66 (n = 93) for 894G and 0.34 (n = 47) for 894T. Genotype frequencies were consistent with those expected under Hardy-Weinberg-Equilibrium (
Female Patients with CF FENO did not differ between premenstrual (mean age 9.2 ± 2.8 years, n = 12) and adult (20.1 ± 8.0 years, n = 18) female patients with CF (4.9 ± 3.6 ppb versus 5.2 ± 3.7 ppb, p = 0.84). Bivariate analysis indicated no significant relationship between FENO levels and  F508 heterozygozity (F1,30 = 1.04, p = 0.31). Bivariate analysis in the female patients with CF indicated significant differences in FENO levels by 894G/T genotype. Mean FENO was significantly higher in the 894T allele carriers than in 894G homozygotes (7.0 ± 4.4 ppb versus 3.6 ± 1.9 ppb, t15 = 2.57, p = 0.02) (Figure 1)
. Higher FENO in individuals with at least one 894T allele was seen in both adult females (7.4 ± 4.5 ppb versus 3.5 ± 1.7 ppb, p = 0.05) and premenstrual patients with CF (6.4 ± 4.8 ppb versus 3.8 ± 2.3 ppb, p = 0.23); although possibly caused by the relatively small number studied, this difference did not reach statistical significance in the younger age group. The mean FEV1 percentage predicted was also higher in the female patients with at least one 894T allele (Table 2); however, this trend did not reach formal statistical significance (t23 = 1.82, p = 0.08). A significant difference was found in sputum microbiology—colonization of airways with P. aeruginosa was more frequent in female homozygous for the wild-type allele when compared with carriers of the 894T allele (p < 0.05) (Table 2).
A generalized linear model of FENO was constructed to investigate the multivariate relationship of age, FEV1, P. aeruginosa positivity in sputum, CFTR F508 heterozygozity, and NOS3 genotype to FENO levels. The results indicated that independently of the other covariates in the model, the presence of at least one 894T allele in the NOS3 gene was significantly associated with increased FENO levels (F1,23 = 5.93, p = 0.02). Although our power to detect significant interactions was low, there was no evidence of a significant interaction between CFTRF508 heterozygosity and NOS3 genotype in this model (data not shown).
In this study, we show that the 894T variant in exon 7 of the NOS3 gene is associated with higher FENO and a decreased risk for colonization of airways with P. aeruginosa in female patients with CF. These data support the hypothesis that naturally occurring variants in constitutive NOS contribute to both variable NO formation and NO-related host defense mechanisms in CF airways. The activity of vascular NOS3 is known to be related to an individual's sex. The higher activity of NOS3 in females compared with males is caused by circulating estrogen, which activates plasma membrane–associated estrogen receptors coupled to NOS3 (15). This sex difference has been shown to be of physiologic importance. For instance, the higher release of NO from NOS3 of the vascular endothelium in females contributes to sex-related differences in coronary artery function (20). An important role of the estrogen–NOS3 interaction has also been shown for the regulation of peripheral vascular resistance. In a study in humans, forearm vascular constrictor response to infusion of the NOS inhibitor N(G)-monomethyl-L-arginine (L-NMMA) was higher in premenopausal women compared with both postmenopausal women and men and was significantly increased in postmenopausal women after estrogen replacement therapy (21). In vitro studies show that estrogen treatment results in a rapid increase in the activity of NOS3 in human airway epithelial cells via nontranscriptional mechanisms (10); however, increased activity of NOS3 in healthy females is not reflected in FENO, as healthy females have lower levels of FENO than males (22). An earlier report suggesting that FENO was related to the menstrual cycle (23) has not been confirmed in subsequent studies (24, 25). The results from our data further suggest that the relationship between FENO and NOS3 genotype is unrelated to effects of estrogen on the activity of NOS3, as (1) FENO did not differ between male and female patients with CF, (2) FENO did not correlate with age in females, and (3) the association between the 894T allele and higher FENO was seen in both prepubertal and adult female patients with CF. The 894G/T NOS3 gene variant has been used as a disease marker in a number of association studies (26–29). The special interest in this marker results from the fact that it is located in exon 7 of the gene and is a nonsynonymous single nucleotide polymorphism coding for either glutamate or aspartate at position 298 in the NOS3 protein. Because glutamate and aspartate are considered conservative substitutions, the observed associations between this polymorphism and certain cardiovascular diseases such as coronary artery disease, myocardial infarction, and essential hypertension (27–29) had not been thought to be causative, but rather to reflect linkage–disequilibrium with a truly functional mutation somewhere else in the NOS3 gene; however, it was recently demonstrated in vitro that the NOS3 protein with aspartate but not glutamate at position 298 is cleaved into a 100-kD fragment and a 35-kD fragment (14), suggesting altered NO formation by the variant protein. Our results showing increased airway NO concentrations in individuals harboring the mutant NOS3 are consistent with findings of an association between the 894T allele and high plasma NO metabolite concentrations in healthy individuals (18). Further support for increased NO formation in biologic systems with deficient NOS3 protein comes from experiments with mice having a targeted deletion of the nos3 gene. These animals had a twofold increase in FENO compared with wild-type animals (11). The association between the 894T allele and higher FENO in female patients with CF could also reflect differences in retention or consumption of NO within the airways. Concentrations of nitrate have been shown to be significantly higher in airway secretions of patients with CF compared with control subjects (6, 30). In addition, a positive correlation was observed between FENO and pulmonary function (4, 31), suggesting that low FENO in patients with CF with more advanced pulmonary disease may either result from reduced NO formation or increased retention and/or consumption of NO within airway secretions. Retention is affected by viscoelastic properties because nitrate concentrations correlate with sputum rigidity (32). Conversion of airway NO into nitrate may result from the reaction of NO with superoxide anion forming peroxynitrite and subsequently nitrate, as recently demonstrated in vitro (33). Furthermore, there is recent evidence that enzymatic NO consumption in the CF airways occurs through NO reductase containing P. aeruginosa strains (34). Higher FENO in female patients with CF with the 894T allele may thus result from increased NO formation or reflect better airway clearance, decreased viscoelasticity of airway secretions, a lower quantity of superoxide-releasing neutrophils, or differences in colonization with denitrifying pathogens.
We have recently established that the size of a gene polymorphism in a different isoform of NOS, namely NOS1, which is also constitutively expressed within the human airways, is associated with higher FENO in both patients with asthma and those with CF (7, 35, 36). In CF, this effect was stronger in the group of In conclusion, our data suggest that a common variant in the NOS3 gene may affect airway NO concentrations, pulmonary function, and susceptibility to respiratory pathogens in female CF patients. Further studies are needed to understand the mechanisms involving genotype-related activation of NOSs and NO-related physiology in the airways.
The authors thank the patients that participated in this study and Ms. Manuela Groch for assistance with the FENO measurements.
Supported by a grant from the Deutsche Forschungsgemeinschaft (K.S.v.G.). Received in original form February 26, 2002; accepted in final form August 11, 2002
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
2 = 0.59, p = 0.74). The 894T allele was significantly more frequent in male than female CF patients (40% versus 21%, p = 0.014) (
Asp) is a major risk factor for coronary artery disease in the UK. Circulation 1999;100:1515–1520.
